Estimating blood loss: comparative study of the accuracy of parents and health care professionals.

OBJECTIVE: Hematemesis and hematochezia are not uncommon presenting complaints in children. The amount of blood loss reported by the parent is likely to influence the pediatrician's decision regarding investigations and management. Currently, there are only very limited data regarding the ability of laypersons to estimate blood losses visually. This study investigated the accuracy of parents, in comparison to pediatric health care professionals, in estimating blood loss volumes. PATIENTS AND METHODS: We performed a prospective, single-blinded study including 227 participants, comprising 131 parents, 58 nurses, and 38 doctors. Participants visually estimated the volume of 1 randomly allocated sample from each of the 2 categories: ( 1) 1, 5, or 10 mL of artificial blood applied to a diaper ( simulated hematochezia) and ( 2) 5, 10, or 50 mL placed in a kidney-dish ( simulated hematemesis). An "error factor" (=, estimated volume/actual volume shown) was used to facilitate comparisons. RESULTS: Parents provided the most inaccurate estimates overall, although individual accuracy varied considerably. The largest overestimate ( 518 mL) and the highest error factor (23.4) were recorded in a parent; overall, 71% of the estimates provided by parents were overestimates. The highest proportion of accurate estimates (+/- 50% of actual volume) was recorded by nurses (29%). Doctors had a tendency to underestimate volumes (62% of the estimates were less than half the actual volume). However, there was no statistically significant difference between the performance of nurses and doctors. Health care professionals tended to overestimate small volumes and underestimate large volumes. Professional experience had no relevant impact on accuracy, nor did parental gender or age. CONCLUSIONS: Visual estimation of blood losses is highly inaccurate, both by laypersons and by health care professionals. Physicians should, therefore, base management decisions primarily on clinical findings and not overly rely on the history provided, or their own estimates. Pediatrics 2009; 124: e729-e736